Display Calibration 101: Step-By-Step With Datacolor's Spyder4Elite

Test Patterns 101: What We Measure

To create a calibration test suite, one needs a method of pattern generation. Most patterns are simple squares of color or gray that the meter reads to determine what adjustments are needed. The accuracy of these patterns is terribly important since measuring the wrong thing will result in a monitor that looks worse than before!

We use a standalone signal generator that sends reference patterns to the display. It’s made by Accupel and is about 1.5 times the price of the X-Rite i7 Pro. While extremely useful in the lab, it isn’t exactly practical for most computer users. The alternative of course, is to use your computer to generate the patterns.

In the case of phones and tablets, we didn’t have a decent solution until recently. Datacolor recently came out with a product called SpyderGallery, which not only generates patterns on iOS- and Android-based devices, but also creates calibration look-up tables for them. While it’s impossible to actually and fully calibrate a smartphone or tablet because the controls simply aren’t there and the OSes are locked down, this method compensates via software, allowing the user to view more accurate images through the SpyderGallery app.

All modern fixed-pixel displays create images in RGB format. That is, each pixel is made up of a red, green, and blue sub-pixel. These sub-pixels are lit to one of 256 luminance levels to render the required color. Thanks to this, it’s easy to create a pattern from simple RGB values. For example, this much red plus that much green plus so much blue equals forest green. In the case of reference patterns, a red primary would be R255, G0, B0, in other words, the maximum red signal without the other two primaries in the mix. Gray patterns are no different, simply input the appropriate RGB values for whatever shade of gray you desire.

Calibration software like CalMAN or Spyder4Elite can generate patterns on the fly for a meter to measure. Automating the process in software is a tremendous time-saver, especially when measuring 30+ patterns for a color saturation test!

The patterns themselves are either fields or windows of a particular color or gray level. For instance, when measuring grayscale, 11 patterns are displayed one after the other, from full black to full white, in 10 percent increments. When measuring color, the patterns are color primaries (red, green, blue) and secondaries (cyan, magenta, yellow) and can vary in saturation and brightness level. We’ll cover those terms in more detail in our next article. All you need to know now is that these patterns are what the meter is measuring.

Now, let’s look more at the software itself and how it turns the measurement data into something we can actually use to improve our display.

Create a new thread in the UK Article comments forum about this subject
This thread is closed for comments
14 comments
Comment from the forums
    Your comment
  • Menigmand
    I still don't understand why we have to buy a $250 device just to get a proper picture on our screens. Why are they not factory calibrated? After all, in these days, the input is digital, so the display data should be 1-to-1 identical to (for example) game RGB data... I'm sure I'm wrong somehow, but about what?
  • mi1ez
    That was actually a genuinely interesting article! Good work Tom's.

    885869 said:
    I still don't understand why we have to buy a $250 device just to get a proper picture on our screens. Why are they not factory calibrated? After all, in these days, the input is digital, so the display data should be 1-to-1 identical to (for example) game RGB data... I'm sure I'm wrong somehow, but about what?


    The problem is twofold:

    Variance in the panels. Each panel would need to be calibrated individually adding time and cost

    The WOW factor. Many manufacturers will ship monitors with the saturation set too high so when you fist set it up it gives you the wow factor. I believe. This second one might not be 100% true and more of a showroom trick.
  • Will P
    A great article, a good introduction on an often overlooked aspect of the PC setup. Every screen I have ever bought has had a ridiculously high brightness out of the box, which I suppose could be put down to a 'Wow factor' for sure... The kit is expensive though, but if it is going to make every screen you own 20% better and last for 5 years it starts to look like a good investment.

    I was wondering if the re-calibration of the cheaper units (such as the Spyder) can be done with the Spyder itself or do you need specialist kit?
  • timbozero
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it.
    There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost.
    Was the author sponsored by DataColor or, just wildly out of touch?
  • mi1ez
    249904 said:
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it. There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost. Was the author sponsored by DataColor or, just wildly out of touch?

    How does a blu-ray calibrate a screen then?
  • timbozero
    142373 said:
    249904 said:
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it. There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost. Was the author sponsored by DataColor or, just wildly out of touch?
    How does a blu-ray calibrate a screen then?


    Is short (for the example I mentioned) , you get several display patterns shown and adjust your screen until they display as they should. You will need to do this a few times for each as some will impact others (you would have to with the Datacolor as well unless it can adjust your monitor automatically afaik). The colours are adjusted the same way but with a filter used by you to help 'understand' the data.

    This is a link to typical testcard used by the Blu-Ray http://www.blu-ray.com/movies/screenshot.php?movieid=16216&position=2
    This is a link to a review of the Blu-Ray on the same site http://www.blu-ray.com/movies/Disney-WOW-World-of-Wonder-Blu-ray/16216/

    Hopefully that will help
  • mi1ez
    249904 said:
    142373 said:
    249904 said:
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it. There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost. Was the author sponsored by DataColor or, just wildly out of touch?
    How does a blu-ray calibrate a screen then?
    Is short (for the example I mentioned) , you get several display patterns shown and adjust your screen until they display as they should. You will need to do this a few times for each as some will impact others (you would have to with the Datacolor as well unless it can adjust your monitor automatically afaik). The colours are adjusted the same way but with a filter used by you to help 'understand' the data. This is a link to typical testcard used by the Blu-Ray http://www.blu-ray.com/movies/screenshot.php?movieid=16216&position=2 This is a link to a review of the Blu-Ray on the same site http://www.blu-ray.com/movies/Disney-WOW-World-of-Wonder-Blu-ray/16216/ Hopefully that will help


    So do you have like a physical card to compare with what's on screen?
  • timbozero
    142373 said:
    249904 said:
    142373 said:
    249904 said:
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it. There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost. Was the author sponsored by DataColor or, just wildly out of touch?
    How does a blu-ray calibrate a screen then?
    Is short (for the example I mentioned) , you get several display patterns shown and adjust your screen until they display as they should. You will need to do this a few times for each as some will impact others (you would have to with the Datacolor as well unless it can adjust your monitor automatically afaik). The colours are adjusted the same way but with a filter used by you to help 'understand' the data. This is a link to typical testcard used by the Blu-Ray http://www.blu-ray.com/movies/screenshot.php?movieid=16216&position=2 This is a link to a review of the Blu-Ray on the same site http://www.blu-ray.com/movies/Disney-WOW-World-of-Wonder-Blu-ray/16216/ Hopefully that will help
    So do you have like a physical card to compare with what's on screen?


    No, the only physical item (other than the disc) is a blue tinted filter which is used for one of the several tests.
    Brightness (as an example) displays a test strip and you adjust the brightness until your screen shows certain details but not others (as instructed by the previous 'chapter' of the 'film').
    There are several levels of testing and reading the review I linked will give you more of an idea.

    I did the setup for TV, projector setup (Datacolor products won't work on projectors as the device blocks the very image you need to measure) and my monitor. The projector wasnt far out and made a small difference, my monitor and TV were 'miles' out and the tuning with the disc worked wonders :)
  • mi1ez
    249904 said:
    142373 said:
    249904 said:
    142373 said:
    249904 said:
    It is IMO ridiculous to suggest the spending of near the cost of a monitor to calibrate it. There are several Blu-ray Discs (Disney World of Wonder for example) that will achieve 80%+ of the levels of calibration for 10% of the cost. Was the author sponsored by DataColor or, just wildly out of touch?
    How does a blu-ray calibrate a screen then?
    Is short (for the example I mentioned) , you get several display patterns shown and adjust your screen until they display as they should. You will need to do this a few times for each as some will impact others (you would have to with the Datacolor as well unless it can adjust your monitor automatically afaik). The colours are adjusted the same way but with a filter used by you to help 'understand' the data. This is a link to typical testcard used by the Blu-Ray http://www.blu-ray.com/movies/screenshot.php?movieid=16216&position=2 This is a link to a review of the Blu-Ray on the same site http://www.blu-ray.com/movies/Disney-WOW-World-of-Wonder-Blu-ray/16216/ Hopefully that will help
    So do you have like a physical card to compare with what's on screen?
    No, the only physical item (other than the disc) is a blue tinted filter which is used for one of the several tests. Brightness (as an example) displays a test strip and you adjust the brightness until your screen shows certain details but not others (as instructed by the previous 'chapter' of the 'film'). There are several levels of testing and reading the review I linked will give you more of an idea. I did the setup for TV, projector setup (Datacolor products won't work on projectors as the device blocks the very image you need to measure) and my monitor. The projector wasnt far out and made a small difference, my monitor and TV were 'miles' out and the tuning with the disc worked wonders :)


    It's very much subjective still, rather than absolute. While I agree for most people this is pointless, if you need it calibrated you need it done properly. I wouldn't surprised if you could rent one of these if you looked hard enough.
  • timbozero
    It is not at all subjective, you need to use it to realise that it is 'absolute' but, it is not as accurate as digital metering (only close to).
  • mi1ez
    249904 said:
    It is not at all subjective, you need to use it to realise that it is 'absolute' but, it is not as accurate as digital metering (only close to).


    If you're using your eyes, then it is subjective. You can't say that everyone's eyes are the same, 20/20 or not.
  • timbozero
    *sigh* Think how you will , I am not here to change your mind.

    The brightness test (for example) is done with a graduated display thus :-
    http://static.highdefview.com/wp-content/uploads/review/wow/wow_brightness_thumb.jpg.pagespeed.ce.JIxKCdd3ek.jpg
    There is only 1 point on the brightness adjustment where the correct graduation will match and the only 'downside' (which you could even remotely call subjective) is that it will not be spot on (as a digital meter might be).

    There is another for contrast etc., etc., etc.. Once they are all completed .....

    Ah hell, try it or don't. It made a massive difference to my cinema room and for only 25$ which is a bargain compared to the digital meter (which was my initial point, that it's overkill to buy a meter)

    Peace, Out :)
  • mi1ez
    249904 said:
    *sigh* Think how you will , I am not here to change your mind. The brightness test (for example) is done with a graduated display thus :- http://static.highdefview.com/wp-content/uploads/review/wow/wow_brightness_thumb.jpg.pagespeed.ce.JIxKCdd3ek.jpg There is only 1 point on the brightness adjustment where the correct graduation will match and the only 'downside' (which you could even remotely call subjective) is that it will not be spot on (as a digital meter might be). There is another for contrast etc., etc., etc.. Once they are all completed ..... Ah hell, try it or don't. It made a massive difference to my cinema room and for only 25$ which is a bargain compared to the digital meter (which was my initial point, that it's overkill to buy a meter) Peace, Out :)


    We'll agree to disagree.
  • Flyfisherman
    Very interesting and much needed article, keep up the good work!

    But there is one thing that comes in mind, about the licence of the S/W.
    Is it possible to install the S/W and the colorimeter on more than one computer, or is it locked to the first computer once one has entered the activation code?

    Because I have several computers at home and really would like to use the equipment to do this calibration on them all.

    I assume that there should not be a problem to calibrate two monitors on the same computer?

    Best Regards from Sweden.